\documentclass[letterpaper,12pt,authoryear]{elsarticle}
\usepackage{AEJM, setspace,graphicx,verbatim,xcolor,lscape,amsmath,amssymb,xfrac}
\usepackage{hyperref}
\usepackage{tikz}
\usepackage{rotating}
\usepackage{array}
\usepackage{caption}
\usepackage{pdflscape}




\captionsetup{font=large}
\usepackage{booktabs, setspace, epstopdf, amsfonts, calc}
\hypersetup{
  colorlinks={true},
  linkcolor={black},
  anchorcolor={black},
  citecolor={black},
  filecolor={black},
  menucolor={black},
  urlcolor={blue},
  pdfauthor = {J Schaller},
  pdfkeywords = {}
  pdfdisplaydoctitle={true},
  pdftitle = {},
  pdfsubject = {},
  pdfpagemode = UseNone
}

\usepackage[authoryear]{natbib}


%\biboptions{longnamesfirst,}

\newcommand{\msymb}[1]{\textrm{\textsl{#1}}}
\def\sym#1{\ifmmode^{#1}\else\(^{#1}\)\fi}

%for referencing footnotes
\makeatletter
\newcommand\footnoteref[1]{\protected@xdef\@thefnmark{\ref{#1}}\@footnotemark}
\makeatother
\interfootnotelinepenalty=10000

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Page Layout
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\setlength{\voffset}{-1in}
\setlength{\paperheight}{11in}
\setlength{\topmargin}{1in}
\setlength{\headheight}{0in}
\setlength{\headsep}{0in}
\setlength{\topskip}{0in}
\setlength{\textheight}{624pt}% 9in - 24pt (for page number)
\setlength{\footskip}{24pt}

\setlength{\paperwidth}{8.5in}
\setlength{\hoffset}{-1in}
\setlength{\textwidth}{6.5in}
\setlength{\oddsidemargin}{1in}
\setlength{\evensidemargin}{1in}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Preferences for float placement
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\renewcommand{\textfraction}{0.05}
\renewcommand{\topfraction}{0.95}
\renewcommand{\bottomfraction}{0.95}
\renewcommand{\floatpagefraction}{0.35}
\setcounter{totalnumber}{5}

\doublespacing
%\onehalfspacing
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Some Paths and Parameters
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\newcommand{\figwidth}{0.47\textwidth}
\newcommand{\figboxwidth}{0.4\textwidth}
\newcommand{\minipagewidth}{0.65\textwidth}
\newcommand{\noteswidth}{0.7\textwidth}

%path setup
\newcommand{\bibPath}{./}

\renewcommand{\footnotesize}{\normalsize}
\usepackage{setspace} % load setspace before footmisc
\usepackage[hang,flushmargin]{footmisc}
\setlength{\footnotesep}{\baselineskip} %use 1.67\baselineskip for a double space
\makeatother
\renewcommand{\footnotelayout}{\doublespacing}


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%
% Main Text
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\begin{document}
	\section*{\textbf{Online Appendix Results for Family Support During Hard Times: Dynamics of Intergenerational Exchange after Adverse Events}}
	\thispagestyle{empty}
	\vspace{5mm}
	\textbf{Jessamyn Schaller}, Claremont McKenna College
	
	\textbf{Chase Eck}, Baylor College of Medicine, IQuESt, and MEDVAMC 

	\appendix
	\setcounter{table}{0}
	\setcounter{figure}{0}
	\renewcommand{\thetable}{A\arabic{table}}
	\renewcommand{\thefigure}{A\arabic{figure}}
	\clearpage
	\newpage
	
\clearpage
\newpage
\setcounter{table}{0}
\setcounter{figure}{0}
\renewcommand{\thetable}{B\arabic{table}}
\renewcommand{\thefigure}{B\arabic{figure}}

\section{Additional Figures and Tables}
\begin{figure}[htbp]
	\centering
	\caption{{Coefficient Plot: Event Study---Parental Household Events and Intergenerational Transfers}}\label{coef_plot1}
	
	\begin{tabular}{c}
		\multicolumn{1}{c}{Panel A: Any Parent-to-Child Transfer} \\
		\includegraphics[width=0.8\textwidth]{../output/coefplot/table3_coefplot_at_pk.pdf} \\
		\multicolumn{1}{c}{Panel B: Any Child-to-Parent Transfer} \\
		\includegraphics[width=0.8\textwidth]{../output/coefplot/table3_coefplot_at_kp.pdf} \\
		\multicolumn{1}{c}{Panel C: Any Child Helps with ADLs} \\
		\includegraphics[width=0.8\textwidth]{../output/coefplot/table3_coefplot_any_helpr.pdf} \\
		\multicolumn{1}{p{16cm}}{\small{Notes: This figure plots event study coefficients from estimation of Equation 1. Each regression includes only households that ever experience a particular shock. Observations are weighted using HRS sample weights. Standard errors are clustered by household and 95 percent confidence intervals are included. Stars indicate significance level after a Holm-Bonferroni adjustment for multiple hypothesis tests. $p< .05 =*, p< .01 =**, p< .001 = ***$.}}
	\end{tabular}
\end{figure}

\clearpage
\newpage

\begin{figure}[htbp]
	\centering
	\caption{Coefficient Plot: Event Study---Parental Household Events and Intergenerational Transfers by Household Wealth}\label{coef_plot2}
	
	\begin{tabular}{c}
		\multicolumn{1}{c}{Panel A: Any Parent-to-Child Transfer} \\
		\includegraphics[width=0.9\textwidth]{../output/coefplot/table3_coefplot_byw_at_pk.pdf} \\
		\multicolumn{1}{c}{Panel B: Any Child-to-Parent Transfer} \\
		\includegraphics[width=0.9\textwidth]{../output/coefplot/table3_coefplot_byw_at_kp.pdf} \\
		\multicolumn{1}{c}{Panel C: Any Child Helps with ADLs} \\
		\includegraphics[width=0.9\textwidth]{../output/coefplot/table3_coefplot_byw_any_helpr.pdf} \\
		\multicolumn{1}{p{16cm}}{\small{Notes: This figure plots event study coefficients from estimation of Equation 1. Households are defined as ``high wealth" if their average pre-event wealth is above the full sample's median lifetime wealth and ``low wealth'' if their average pre-event wealth is below the full sample's median lifetime wealth. Each regression includes only households that ever experience a particular shock. Observations are weighted using HRS sample weights. Standard errors are clustered by household and 95 percent confidence intervals are included.}}
	\end{tabular}
\end{figure}


\clearpage
\newpage

\begin{figure}[htbp]
	\centering
	\caption{Costly Health Shocks and Parent-to-Child Financial Transfers by Medicare Eligibility at Time of Shock (ages 55-74 only)}\label{heterog_fig1}
	
	\begin{tabular}{cc}
		\includegraphics[width=0.5\textwidth]{../output/byage/any_rhosp_at_pk_byage.pdf} &
		\includegraphics[width=0.5\textwidth]{../output/byage/any_health3to5_at_pk_byage.pdf}	\\
		\includegraphics[width=0.5\textwidth]{../output/byage/any_rstrok_at_pk_byage.pdf} &
		\includegraphics[width=0.5\textwidth]{../output/byage/any_rcancr_at_pk_byage.pdf}	\\
		\multicolumn{2}{p{16.5cm}}{\small{Notes: This figure shows event study coefficients and 95 percent confidence intervals from estimation of Equation 1. Each figure presents a different treatment variable. Standard errors are clustered by household. Observations are weighted using HRS sample weights.}} \\
	\end{tabular}
\end{figure}

\clearpage
\newpage

\begin{figure}[htbp]
	\centering
	\caption{Specific Health Shocks and Informal Care---Heterogeneity}\label{het_fig2}
	
	\begin{tabular}{cccc}
		\multicolumn{4}{c}{Panel A: By Child Gender} \\
		\includegraphics[width=0.23\textwidth]{../output/bykgender/any_rcardiac_khlp_bykgender.pdf} &
		\includegraphics[width=0.23\textwidth]{../output/bykgender/any_rstrok_khlp_bykgender.pdf} &
		\includegraphics[width=0.23\textwidth]{../output/bykgender/any_rmemory_khlp_bykgender.pdf} &
		\includegraphics[width=0.23\textwidth]{../output/bykgender/any_rcancr_khlp_bykgender.pdf} \\
		\multicolumn{4}{c}{Panel B: By Household Wealth} \\
		\includegraphics[width=0.23\textwidth]{../output/byhhwealth/any_rcardiac_any_helpr_byw.pdf} &
		\includegraphics[width=0.23\textwidth]{../output/byhhwealth/any_rstrok_any_helpr_byw.pdf} &
		\includegraphics[width=0.23\textwidth]{../output/byhhwealth/any_rmemory_any_helpr_byw.pdf}	&
		\includegraphics[width=0.23\textwidth]{../output/byhhwealth/any_rcancr_any_helpr_byw.pdf}	\\
		\multicolumn{4}{c}{Panel C: By Medicare Eligibility at Time of Shock} \\
		\includegraphics[width=0.23\textwidth]{../output/byage/any_rcardiac_any_helpr_byage.pdf} &
		\includegraphics[width=0.23\textwidth]{../output/byage/any_rstrok_any_helpr_byage.pdf} &
		\includegraphics[width=0.23\textwidth]{../output/byage/any_rmemory_any_helpr_byage.pdf}	&
		\includegraphics[width=0.23\textwidth]{../output/byage/any_rcancr_any_helpr_byage.pdf}	\\
		\multicolumn{4}{c}{Panel D: By Household Structure at Time of Shock} \\
		\includegraphics[width=0.23\textwidth]{../output/byhhstructure/any_rcardiac_any_helpr_byhhstructure.pdf} &
		\includegraphics[width=0.23\textwidth]{../output/byhhstructure/any_rstrok_any_helpr_byhhstructure.pdf} &
		\includegraphics[width=0.23\textwidth]{../output/byhhstructure/any_rmemory_any_helpr_byhhstructure.pdf} &
		\includegraphics[width=0.23\textwidth]{../output/byhhstructure/any_rcancr_any_helpr_byhhstructure.pdf} \\
		\multicolumn{4}{p{16cm}}{\small{Notes: This table presents event study coefficients from estimation of Equation 1. The estimation was performed separately by household wealth status. Households are defined as ``high wealth" if their average pre-event wealth is above the full sample's median lifetime wealth and ``low wealth'' if their average pre-event wealth is below the full sample's median lifetime wealth. Each regression includes only households that ever experience a particular shock. Observations are weighted using HRS sample weights.}}
	\end{tabular}
\end{figure}


%OLS REGRESSIONS
\newpage
\begin{table}[htbp]
	\caption{Pooled Regressions}  \label{OLSFE}
	\begin{center}
		\begin{tabular}{lrrrrrr} \toprule
			& \multicolumn{2}{c}{Parent-to-Child } & \multicolumn{2}{c}{Child-to-Parent} & \multicolumn{2}{c}{Child }  \\
			& \multicolumn{2}{c}{Financial Transfer } & \multicolumn{2}{c}{Financial Transfer} & \multicolumn{2}{c}{ Providing Care}  \\
			\cmidrule(l r){2-3}  \cmidrule(l r){4-5} \cmidrule(l r){6-7}
			\input{../output/tableb1/CS_FE.tex}
			\hline
			Household Fixed Effects & No & Yes & No & Yes & No & Yes \\
			\bottomrule
			\multicolumn{7}{p{16cm}}{\small{Notes: All regressions include a quadratic in age and dummies for high school graduate, college attendance, non-white respondent or spouse, number of children (1, 2, 3 or more), and HRS wave. Wealth and income are normalized to 2017 dollars. Observations are weighted using HRS sample weights. Standard errors are clustered at the household level. Significance at 1\% and 5\%, levels are indicated by ** and *, respectively.} } \\
		\end{tabular}
	\end{center}
\end{table}



\clearpage
\newpage
\setcounter{table}{0}
\setcounter{figure}{0}
\renewcommand{\thetable}{C\arabic{table}}
\renewcommand{\thefigure}{C\arabic{figure}}
\section{Robustness Checks}\label{sec:robust}
\begin{figure}[htbp]
	\centering
	\caption{Event Studies: Financial Shocks and Intergenerational Financial Transfers - Borusyak Event Plots}\label{EventStudyA_bor}
	
	\begin{tabular}{cc}
		
		\multicolumn{2}{c}{Panel A: Wealth Loss} \\
		\includegraphics[width=0.45\textwidth]{../output/figurec123/wneg2_at_pk_bor} &
		\includegraphics[width=0.45\textwidth]{../output/figurec123/wneg2_at_kp_bor} 
		\\
		\multicolumn{2}{c}{Panel B: Job Exit} \\
		\includegraphics[width=0.45\textwidth]{../output/figurec123/any_rlost_job_at_pk_bor} &
		\includegraphics[width=0.45\textwidth]{../output/figurec123/any_rlost_job_at_kp_bor} 
		\\
		\multicolumn{2}{p{16cm}}{{Notes: This table presents event study coefficients from estimation of Equation 1 using the imputation estimator developed by \cite{borusyak2021revisiting}. Observations are weighted using HRS sample weights.}}
	\end{tabular}
\end{figure}




\clearpage
\newpage



\begin{figure}[htbp]
	\centering
	\caption{Event Studies: Fatal Health Shocks and Intergenerational Transfers - Borusyak Event Plots}\label{EventStudyB_bor}
	
	\begin{tabular}{ccc}
		\multicolumn{3}{c}{Panel A: Male Widowed} \\
		\includegraphics[width=0.33\textwidth]{../output/figurec123/male_rwidowed_at_pk_bor} &
		\includegraphics[width=0.33\textwidth]{../output/figurec123/male_rwidowed_at_kp_bor} &
		\includegraphics[width=0.33\textwidth]{../output/figurec123/male_rwidowed_any_helpr_bor}
		\\
		\multicolumn{3}{c}{Panel B: Female Widowed} \\
		\includegraphics[width=0.33\textwidth]{../output/figurec123/fem_rwidowed_at_pk_bor} &
		\includegraphics[width=0.33\textwidth]{../output/figurec123/fem_rwidowed_at_kp_bor} &
		\includegraphics[width=0.33\textwidth]{../output/figurec123/fem_rwidowed_any_helpr_bor}
		\\
		\multicolumn{3}{p{16cm}}{{Notes: This table presents event study coefficients from estimation of Equation 1 using the imputation estimator developd by \cite{borusyak2021revisiting}. Observations are weighted using HRS sample weights.}}
	\end{tabular}
\end{figure}





\begin{figure}[htbp]
	\centering
	\caption{Event Studies: Non-Fatal Health Shocks and Intergenerational Transfers - Borusyak Event Plots}\label{EventStudyC_bor}
	
	\begin{tabular}{ccc}
		\multicolumn{3}{c}{Panel A: Hospitalization} \\
		\includegraphics[width=0.33\textwidth]{../output/figurec123/any_rhosp_at_pk_bor} &
		\includegraphics[width=0.33\textwidth]{../output/figurec123/any_rhosp_at_kp_bor}  &
		\includegraphics[width=0.33\textwidth]{../output/figurec123/any_rhosp_any_helpr_bor} \\
		\multicolumn{3}{c}{Panel B: Disability} \\
		\includegraphics[width=0.33\textwidth]{../output/figurec123/any_disab_at_pk_bor} &
		\includegraphics[width=0.33\textwidth]{../output/figurec123/any_disab_at_kp_bor}  &
		\includegraphics[width=0.33\textwidth]{../output/figurec123/any_disab_any_helpr_bor} \\
		\multicolumn{3}{c}{Panel C: Poor Health} \\
		\includegraphics[width=0.33\textwidth]{../output/figurec123/any_health3to5_at_pk_bor} &
		\includegraphics[width=0.33\textwidth]{../output/figurec123/any_health3to5_at_kp_bor}  &
		\includegraphics[width=0.33\textwidth]{../output/figurec123/any_health3to5_any_helpr_bor} \\
		\multicolumn{3}{p{16cm}}{{Notes: This table presents event study coefficients from estimation of Equation 1 using the imputation estimator developd by \cite{borusyak2021revisiting}. Observations are weighted using HRS sample weights.}}
	\end{tabular}
\end{figure}

\clearpage
\newpage



\begin{table}[htbp]
	\caption{Alternative Wealth Shocks}  \label{altws}
	\begin{center}
		\begin{tabular}{lcccc} \toprule
			& No     & No Nursing &   Non-Housing  &  Financial       \\
			Event:			& Imputations   & Home Change  &  Wealth Loss & Wealth Loss     \\
			\toprule
			\multicolumn{5}{c}{Panel A: Any Parent-to-Child Transfer} \\
			\input{../output/tablec1/at_pk_post_ws.tex}
			\bottomrule
			\multicolumn{5}{c}{Panel B: Any Child-to-Parent Transfer} \\
			\input{../output/tablec1/at_kp_post_ws.tex}
			\bottomrule
			\multicolumn{5}{c}{Panel C: Any Child Helps with ADLs} \\
			\input{../output/tablec1/any_helpr_post2_ws.tex} 
			\bottomrule
			\multicolumn{5}{p{12cm}}{{Notes: This table presents event study coefficients from estimation of Equation 1. Each regression includes only households that ever experience a particular shock. Observations are weighted using HRS sample weights. Standard errors are clustered by household. Significance at 1\% and 5\% levels are indicated by ** and *, respectively.}}
		\end{tabular}
	\end{center}
\end{table}

\begin{table}[htbp]
	\caption{Alternative Job Exit Measures}  \label{altjl}
	\begin{center}
		\begin{tabular}{lcccc} \toprule
			& Male    & Female &   Involuntary  &  Health-Related       \\
			Event:			& Job  Exit   & Job Exit  &  Job Loss & Job Exit     \\
			\toprule
			\multicolumn{5}{c}{Panel A: Any Parent-to-Child Transfer} \\
			\input{../output/tablec2/at_pk_post_jl.tex}
			\bottomrule
			\multicolumn{5}{c}{Panel B: Any Child-to-Parent Transfer} \\
			\input{../output/tablec2/at_kp_post_jl.tex}
			\bottomrule
			\multicolumn{5}{c}{Panel C: Any Child Helps with ADLs} \\
			\input{../output/tablec2/any_helpr_post2_jl.tex}
			\bottomrule
			\multicolumn{5}{p{11cm}}{{Notes: This table presents event study coefficients from estimation of Equation 1. Each regression includes only households that ever experience a particular shock. Observations are weighted using HRS sample weights. Standard errors are clustered by household. Significance at 1\% and 5\% levels are indicated by ** and *, respectively.}}
		\end{tabular}
	\end{center}
\end{table}



\begin{table}[htbp]
	\caption{Controlling for Pre-Trends}  \label{eq2results}
	\begin{center}
		\begin{tabular}{lccccccc} \toprule
			& Wealth    & Job &   Male  &  Female &  Any  &  Disability & Poor    \\
			Event:			&  loss   & exit  &  widowed & widowed  & hosp & onset & health   \\
			\toprule
			\multicolumn{8}{c}{Panel A: Any Parent-to-Child Transfer} \\
			\input{../output/tablec3/at_pk_post_eq2.tex}
			\bottomrule
			\multicolumn{8}{c}{Panel B: Any Child-to-Parent Transfer} \\
			\input{../output/tablec3/at_kp_post_eq2.tex}
			\bottomrule
			\multicolumn{8}{c}{Panel C: Any Child Helps with ADLs} \\
			\input{../output/tablec3/any_helpr_post_eq2.tex}
			\bottomrule
			\multicolumn{8}{p{15.5cm}}{{Notes: In this model, we estimate a semi-parametric model that controls for pre-event trends, as in \cite{Dobkin_2018}. Specifically, these results are from estimating Equation \ref{eq2}.}}
		\end{tabular}
	\end{center}
\end{table}

\begin{table}[htbp]
	\caption{Unweighted Regressions}  \label{unw}
	\begin{center}
		\begin{tabular}{lccccccc} \toprule
			& Wealth    & Job &   Male  &  Female &  Any  &  Disability & Poor    \\
			Event:			&  loss   & exit  &  widowed & widowed  & hosp & onset & health   \\
			\toprule
			\multicolumn{8}{c}{Panel A: Any Parent-to-Child Transfer} \\
			\input{../output/tablec4/at_pk_post_unw.tex}
			\bottomrule
			\multicolumn{8}{c}{Panel B: Any Child-to-Parent Transfer} \\
			\input{../output/tablec4/at_kp_post_unw.tex}
			\bottomrule
			\multicolumn{8}{c}{Panel C: Any Child Helps with ADLs} \\
			\input{../output/tablec4/any_helpr_post_unw.tex}
			\bottomrule
			\multicolumn{8}{p{15.5cm}}{{Notes: This table presents event study coefficients from estimation of Equation 1. Each regression includes only households that ever experience a particular shock. Observations are unweighted. Standard errors are clustered by household. Significance at 1\% and 5\% levels are indicated by ** and *, respectively.}}
		\end{tabular}
	\end{center}
\end{table}



\begin{table}[htbp]
	\caption{Adding a Control Group of Never-Treated Households}  \label{cntrl}
	\begin{center}
		\begin{tabular}{lccccccc} \toprule
			& Wealth    & Job &   Male  &  Female &  Any  &  Disability & Poor    \\
			Event:			&  loss   & exit  &  widowed & widowed  & hosp & onset & health   \\
			\toprule
			\multicolumn{8}{c}{Panel A: Any Parent-to-Child Transfer} \\
			\input{../output/tablec5/at_pk_post_cntrl.tex}
			\bottomrule
			\multicolumn{8}{c}{Panel B: Any Child-to-Parent Transfer} \\
			\input{../output/tablec5/at_kp_post_cntrl.tex}
			\bottomrule
			\multicolumn{8}{c}{Panel C: Any Child Helps with ADLs} \\
			\input{../output/tablec5/any_helpr_post_cntrl.tex}
			\bottomrule
			\multicolumn{8}{p{15.5cm}}{{Notes: This table presents event study coefficients from estimation of Equation 1. Each regression includes all households in the sample, including those that do not experience a shock. Observations are weighted using HRS sample weights. Standard errors are clustered by household. Significance at 1\% and 5\% levels are indicated by ** and *, respectively.}}
		\end{tabular}
	\end{center}
\end{table}

\begin{table}[htbp]
	\caption{Borusyak et al. Imputation}  \label{bor_table}
	\begin{center}
		\begin{tabular}{lccccccc} \toprule
			& Wealth    & Job &   Male  &  Female &  Any  &  Disability & Poor    \\
			Event:			&  loss   & exit  &  widowed & widowed  & hosp & onset & health   \\
			\toprule
			\multicolumn{8}{c}{Panel A: Any Parent-to-Child Transfer} \\
			\input{../output/tablec6/at_pk_bor3.tex}
			\bottomrule
			\multicolumn{8}{c}{Panel B: Any Child-to-Parent Transfer} \\
			\input{../output/tablec6/at_kp_bor3.tex}
			\bottomrule
			\multicolumn{8}{c}{Panel C: Any Child Helps with ADLs} \\
			\input{../output/tablec6/any_helpr_bor3.tex}
			\bottomrule
			\multicolumn{8}{p{15.5cm}}{{Notes: This table presents event study coefficients from estimation of Equation 1 using the imputation estimator developed by \cite{borusyak2021revisiting}. Observations are weighted using HRS sample weights. Standard errors are clustered by household. Significance at 1\% and 5\% levels are indicated by ** and *, respectively.}}
		\end{tabular}			
	\end{center}
\end{table}


\begin{table}[htbp]
	\caption{No Event Restrictions and Allowing for Multiple Events}  \label{me}
	\begin{center}
		\begin{tabular}{lccccccc} \toprule
			& Wealth    & Job &   Male  &  Female &  Any  &  Disability & Poor    \\
			Event:			&  loss   & exit  &  widowed & widowed  & hosp & onset & health   \\
			\toprule
			\multicolumn{8}{c}{Panel A: Any Parent-to-Child Transfer} \\
			\input{../output/tablec7/at_pk_post_me.tex}
			\bottomrule
			\multicolumn{8}{c}{Panel B: Any Child-to-Parent Transfer} \\
			\input{../output/tablec7/at_kp_post_me.tex}
			\bottomrule
			\multicolumn{8}{c}{Panel C: Any Child Helps with ADLs} \\
			\input{../output/tablec7/any_helpr_post_me.tex}
			\bottomrule
			\multicolumn{8}{p{16cm}}{{Notes: This table shows results from regression estimates in which multiple events are allowed for each event-type and household (i.e. repeated hospitalizations), and no restrictions are placed on the events. For these models, we do not require that there be two waves observed with no event prior to the event and we allow multiple event indicators  (of the same type) to be ``turned on" simultaneously. For more on estimating event studies when multiple events are possible, see \cite{sandler2014multiple, krolikowski2018choosing}.}}
		\end{tabular}
	\end{center}
\end{table}

\clearpage
\newpage
\begin{table}[htbp]
	\caption{Health Shocks and Informal Care---Heterogeneity}  \label{het_ic_se}
	\begin{center}
		\begin{tabular}{lcccccccc} \toprule
			Subsample:	& Female    & Male  &   &  Not  &  Under & Over & Low &  High  \\
						& Children    & Children &  Single  &  Single  &  Age 65 & Age 65 & Wealth &  Wealth\\
			\toprule
			\multicolumn{9}{c}{Panel A: Any Hospitalization} \\
			\bottomrule
			\input{../output/tablec8/a_h_post2_any_rhosp.tex}
			\bottomrule
			\multicolumn{9}{c}{Panel B: Any Disability} \\
			\bottomrule
			\input{../output/tablec8/a_h_post2_any_disab.tex}
			\bottomrule
			\multicolumn{9}{c}{Panel C: Any Poor Health} \\
			\bottomrule
			\input{../output/tablec8/a_h_post2_any_health3to5.tex}
			\bottomrule
			\multicolumn{9}{p{17cm}}{{Notes: This table presents event study coefficients from estimation of Equation 1. Each regression includes only households that ever experience a particular shock, denoted by the panel; each column is a different subsample of the data. The outcome for each regression is receipt of informal care. Observations are weighted using HRS sample weights. Standard errors are clustered by household. Significance at 1\% and 5\% levels are indicated by ** and *, respectively.}}
		\end{tabular}
	\end{center}
\end{table}

\clearpage
\newpage

\begin{table}[htbp]
	\caption{Specific Health Shocks and Informal Care---Heterogeneity}  \label{het_ic_sc_se}
	\begin{center}
		\begin{tabular}{lcccccccc} \toprule
			Subsample:	& Female    & Male  &   &  Not  &  Under & Over & Low &  High  \\
& Children    & Children & Householder  &  Single  &  Age 65 & Age 65 & Wealth &  Wealth\\
			\toprule
			\multicolumn{9}{c}{Panel A: Any Cardiac} \\
			\bottomrule
			\input{../output/tablec9/a_h_post2_any_rcardiac.tex}
			\bottomrule
			\multicolumn{9}{c}{Panel B: Any Stroke} \\
			\bottomrule
			\input{../output/tablec9/a_h_post2_any_rstrok.tex}
			\bottomrule
			\multicolumn{9}{c}{Panel C: Any Memory} \\
			\bottomrule
			\input{../output/tablec9/a_h_post2_any_rmemory.tex}
			\bottomrule
			\multicolumn{9}{c}{Panel D: Any Cancer} \\
			\bottomrule
			\input{../output/tablec9/a_h_post2_any_rcancr.tex}
			\bottomrule
			\multicolumn{9}{p{17cm}}{{Notes: This table presents event study coefficients from estimation of Equation 1. Each regression includes only households that ever experience a particular shock. Each regression includes only households that ever experience a particular shock, denoted by the panel; each column is a different subsample of the data. The outcome for each regression is receipt of informal care. Observations are weighted using HRS sample weights. Standard errors are clustered by household. Significance at 1\% and 5\% levels are indicated by ** and *, respectively.}}
		\end{tabular}
	\end{center}
\end{table}
 
\end{document}